The present study aims to explore the efficacy and safety of DBS targeted the combination of the anterior limb of internal capsule and nucleus accumbens (ALIC/NAcc-DBS) to refractory OCD, based on multicenter, randomized, double-blinded, sham-stimulation controlled design, that could help to obtain stronger evidence of the efficacy of multiple targets.

OCD is a chronic and disabled mental disorder, with one third of patients could hardly get recovery from the medication and psychotherapy. Deep brain stimulation (DBS) is a novel potential surgical treatment for refractory OCD, however, the choose of targets are still controversial. This study aims to investigate the therapeutic efficacy and safety of bilateral DBS of the anterior limb of internal capsule and nucleus accumbens (ALIC/NAcc-DBS) for treatment-refractory OCD. In this multicenter, prospective, and double-blinded study, 64 patients with OCD who met the eligibility criteria will be enrolled and randomized allocated to active and sham-stimulation group, followed up for three months after surgery. Then the blind will be uncover, the sham group switch to true stimulation, all patients will be followed up at six months after surgery. The primary outcome was to compare the treatment response rate between the active group and sham-controlled group at the end of three months after surgery. The secondary outcomes include changes in the Yale-brown obsessive-compulsive scale (Y-BOCS), Hamilton anxiety scale (HAMA), Hamilton depression scale (HAMD), and Clinical global impressions (CGI) scale of both groups.

Inclusion Criteria:

• aged 18-65 years old;
• a diagnosis of OCD based on The Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5);
• suffered from severe OCD symptoms over 5 years with substantial functional impairment;
• the Y-BOCS total score ≥ 25;
• met the criteria of treatment refractory. Treatment refractory is defined as failed 1) at least three adequate therapeutic trials of serotin reuptake inhibitors (SRIs), with clomipramine being one of the SRIs trials, 2) the use of at least two atypical antipsychotics as augmenting agents, 3) at least 12 sessions of therapist-guided cognitive-behavioral therapy while on a therapeutic dose of an SRIs. These patients have "failed" treatment by demonstrating <25% reduction of Y-BOCS scores or, despite >25% reduction in YBOCS score, by still experiencing significant impairment from their illness.

Exclusion Criteria:

• a history of current or past diagnosis of any psychosis according to DSM-5 except for OCD
• any clinical significant neurological disorder, traumatic brain injury, dementia, or medical illness
• any contradiction of neurosurgery
• any current or unstably remitted substance abuse or dependence
• women who are pregnant or preparing for pregnant
• severe suicide risk and tendency according to the investigators' judgements
• participate in any other clinical trials within three months (except for registered studies).

Randomization of DBS are blinded to evaluators, participants, and programming doctors. The programming doctors used the specific software to allocated the participants, that software was preset random coding program to randomized allocate the subjects. The DBS interface of all patients is displayed as a true stimulus, but the true impulse was running in the background controlled by the software, so that even the programming doctors could not know the grouping.